The broad aim of the proposed research is to explore how information is represented in the human visual system, and in particular to test conditions under which the visual system encodes information relative to a norm or neutral response level. This principle is central to models of color vision (where color is encoded relative to a neutral white point) and has recently been proposed to account for the visual representation of shapes and faces. Norm-based codes make specific and testable predictions about how sensitivity and perception should change when observers are adapted to a stimulus. I will use psychophysical measurements of these adaptation effects to address three specific aims. First, norm-based codes assume that stimuli that appear neutral (e.g. gray) reflect a neutral point in the visual code, and I will test whether there is a necessary correspondence between perceptual norms and response norms, and the potential sites within the visual system at which this relationship is established. Second, I will test whether the principles of adaptation and renormalization in color coding are similar for the encoding of spatial information. Third, I will test whether the principle of renormalization can account for perceptual constancy despite variations in visual sensitivity either across space (at different retinal locations) or across time (as the visual system ages). This work will allow me to characterize normal visual function and how it changes with aging. Relevance: the proposed research will help reveal important design principles in the human visual system and how the visual system is calibrated to compensate for variations in sensitivity or changes in vision with aging. Understanding normal vision provides an important baseline for assessing the consequences of visual disease. [unreadable] [unreadable] [unreadable]